The Parkland grading scale for cholecystitis

Accepted Manuscript The Parkland grading scale for cholecystitis Tarik D. Madni, David E. Leshikar, Christian T. Minshall, Paul A. Nakonezny, Canon C. Cornelius, Jonathan B. Imran, Audra T. Clark, Brian H. Williams, Alexander L. Eastman, Joseph P. Minei, Herb A. Phelan, Michael W. Cripps PII:

S0002-9610(17)30565-2

DOI:

10.1016/j.amjsurg.2017.05.017

Reference:

AJS 12392

To appear in:

The American Journal of Surgery

Received Date: 22 March 2017 Revised Date:

18 May 2017

Accepted Date: 29 May 2017

Please cite this article as: Madni TD, Leshikar DE, Minshall CT, Nakonezny PA, Cornelius CC, Imran JB, Clark AT, Williams BH, Eastman AL, Minei JP, Phelan HA, Cripps MW, The Parkland grading scale for cholecystitis, The American Journal of Surgery (2017), doi: 10.1016/j.amjsurg.2017.05.017. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ABSTRACT Background Gallbladders (GBs) with severe inflammation have longer operative times and an increased risk

RI PT

for complications. We propose a grading system using intraoperative images to better stratify GB inflammation. Methods

SC

After reviewing the intraoperative images of GBs obtained during several hundred laparoscopic cholecystectomies, we developed a five-tiered grading system based on anatomy and

M AN U

inflammatory changes. Fifty intraoperative photographs were taken prior to dissection and then distributed to 11 surgeons who rated each GB’s severity per the grading system. The two-way random effects Intraclass Correlation Coefficient (ICC) was used to assess the reliability among the raters.

TE D

Results

The ICC among the raters of GB severity was 0.804 (95% CI: 0.733 to 0.867; p = 0.0001). Nineteen GB images had greater than 82% agreement and 16 were clustered around GBs with

Conclusion

EP

severe inflammation (grades 3-5).

AC C

This study proposes a simple, reliable grading system that characterizes GB complexity based on inflammation and anatomy.

ACCEPTED MANUSCRIPT

KEWORDS

AC C

EP

TE D

M AN U

SC

RI PT

Gallbladder, grading scale, inflammation, outcomes, reimbursement

ACCEPTED MANUSCRIPT 1

The Parkland Grading Scale for Cholecystitis

2. David E. Leshikar, MDb [email protected] 3. Christian T. Minshall, MD, PhDc [email protected]

5. Canon C. Cornelius, BSc [email protected]

M AN U

4. Paul A. Nakonezny, PhDd [email protected]

SC

Authors 1. Tarik D. Madni, MDa [email protected]

RI PT

Brief title: Cholecystitis Grading Scale

6. Jonathan B. Imran, MDa [email protected]

TE D

7. Audra T. Clark, MDa [email protected]

8. Brian H. Williams, MDa [email protected]

EP

9. Alexander L. Eastman, MD, MPHc [email protected]

AC C

10. Joseph P. Minei, MD, MBAc [email protected] 11. Herb A. Phelan, MD, MSCSc [email protected] 12. Michael W. Cripps, MDc [email protected] a

UT Southwestern Department of Surgery; Dallas, TX UC Davis; Division of Trauma and Acute Care Surgery; Sacramento, CA c UT Southwestern Division of Burn/Trauma/Critical Care; Dallas, TX b

ACCEPTED MANUSCRIPT 2

UT Southwestern Department of Clinical Sciences, Division of Biostatistics; Dallas, TX

SC

Corresponding author Michael W. Cripps, MD University of Texas Southwestern Medical Center Division of Burn/Trauma/Critical Care 5323 Harry Hines Blvd. Dallas, Texas 75390-9158 Office: 214-648-2708 Fax: 214-648-5477 [email protected]

RI PT

d

AC C

EP

TE D

M AN U

Meeting submitted to: This original work was presented by poster at the 2017 annual meeting of EAST and has not been submitted or published elsewhere.

ACCEPTED MANUSCRIPT 3

INTRODUCTION Gallbladder (GB) disease affects over 20 million people in the United States,1 making laparoscopic cholecystectomy (LC) one of the most common operations performed by general

RI PT

surgeons.2 Yet, not all cholecystitis is created equal. Differences in anatomy and inflammation can wreak havoc on an otherwise straightforward operation. Increased degrees of GB

inflammation have been shown to lead to more open conversion and iatrogenic injuries.3

SC

However, outcome comparisons, surgeon compensation, and resident case log entry all consider

rare find for acute care surgeons.

M AN U

LC at the lowest common denominator of a straightforward “robin’s egg blue” GB, which is a

Accurate and reliable stratification of the severity of GB disease requires a grading system that can be widely deployed and easily implemented. Multiple grading scales have been developed in the past to try to predict the level of difficulty for LC. Most of these scores are

TE D

based on preoperative clinical findings, with few that utilize intraoperative factors.4 These scoring systems are also complex, with multiple inputs and grades, limiting the practicality of using these scores in the operative setting. We propose that while preoperative indicators may

EP

hold some predictive value, it is not until the GB is visualized during surgery that a true determination can be made as to the severity of inflammation. To date, there are no widespread,

AC C

validated grading systems utilized to stratify the intraoperative severity of GB inflammation. We believe that this is due, in part, to the inability to routinely capture high-resolution images that can be reviewed and graded by multiple surgeons. Recent improvements in technology that allow for image capture into the patient’s electronic medical record provide the opportunity for more than one surgeon to view and evaluate the severity of disease.

ACCEPTED MANUSCRIPT 4

We hypothesize that a novel grading system, based solely on intraoperative images, can stratify GB inflammation. Our first specific aim was to develop a cholecystitis stratification system with a limited number of grades based on intraoperative images. Our second specific aim

RI PT

was to evaluate the reliability of cholecystitis severity grade assignment among a group of acute care surgeons.

SC

METHODS Study Setting and Procedures

M AN U

This study was approved by the institutional review board at the University of Texas Southwestern Medical Center. All patients at Parkland Memorial Hospital, a large, urban, tertiary referral hospital, who underwent LC by the Trauma and Acute Care Surgery service over an 8month period between 10/2015 and 5/2016 were included in the study. Since October 2015, the

TE D

standard operating procedure for LC at Parkland Memorial Hospital is to take intraoperative pictures of the GB using the laparoscope once the GB is initially visualized. These pictures were stored on the Parkland Black Diamond Video server. Intraoperative photographs were taken of

EP

the right upper abdominal structures after placement of all four laparoscopic ports. If the GB was visualized easily, it was grasped and retracted cephalad prior to taking the photograph. If severe

AC C

inflammation was present which limited mobilization or the ability to visualize the GB, the picture was taken of the inflamed area. These images were referred to as the “initial view” of the GB.

Development of the grading scale was based on the concept that it should have 1) a

limited number of grades, 2) be easy to remember, 3) and have consistent assignment among users. The grading scale was developed by reviewing downloaded initial view images of the

ACCEPTED MANUSCRIPT 5

GBs. We aimed for an odd-numbered grading system but felt three grades of severity (ie: none, moderate, severe) would be too broad of distinctions. Additionally, a seven-tiered grading system was thought to offer too much specificity, would be difficult to remember, and would be

RI PT

cumbersome to use in clinical practice. Our hypothesis was that a five-tiered grading system would give an appropriate, stepwise range for surgeons to differentiate one gallbladder from

SC

another in terms of severity.

Statistical Analysis

M AN U

We first determined the number of photographs and raters needed to achieve at least 80% statistical power, with a significance level of 0.05 (2-tailed), so as to detect a conservative intraclass correlation coefficient (ICC) ≤ 0.20. Thus, an a priori power analysis determined that 11 raters grading 45 GB images achieves 81% power to detect ICC as small as 0.20 with a

TE D

significance level of 0.05. To err on the side of improved power, an independent reviewer selected 50 images equally representative of all five grades and placed them in random order. Afterwards, these 50 intraoperative photographs were distributed (in random order) to 11 acute

EP

care surgeons who, in turn, rated each GB’s severity per the grading system. The two-way random effects ICC was used to assess the reliability (or magnitude of absolute agreement)

AC C

among the 11 raters based upon the 50 GB images.5 Demographic and clinical characteristics of the 50 patients associated with the 50 respective intraoperative photographs across the 5 image grades were described using the sample median and interquartile range (IQR) for continuous variables and the frequency and percentage for categorical variables. Statistical analysis was carried out using SAS software, version 9.4 (SAS Institute, Cary, NC). The level of significance was set at α = .05 (two-tailed).

ACCEPTED MANUSCRIPT 6

RESULTS Over 200 “initial view” images of GBs were available for analysis. A five-tiered grading

RI PT

scale was developed by a team of acute care surgeons that described increasing levels of GB inflammation and commonly encountered anatomic abnormalities. Once developed, an

independent evaluator reviewed the 200 images and selected a cohort of approximately 30

SC

images to test the scale among a group of acute care surgeons. None of these images were used in the final grading scale analysis. After review and discussion of these images, a final grading

each grade noted in Figure 1.

M AN U

scale based on the initial view was developed and the results are in Table 1, with examples of

After the 50 initial view images were randomly ordered, then each of the 11 raters graded all 50 images, for a total of 550 observations. The ICC among the 11 raters of GB severity was

TE D

0.804 (95% CI: 0.733 to 0.867; p = 0.0001; Figure 2). The high reliability of the grading system is also demonstrated by the near-identical trend for all 11 raters, as shown in Figure 2. Nineteen GB images had greater than 82% agreement, and 16 of those were clustered around those GBs

EP

with severe inflammation (grades 3-5). Only 3 images had less than 50% agreement and those were assigned grades 1 through 4.

AC C

We then grouped the GB images by their pathologic diagnosis of 1) cholelithiasis only, 2)

acute cholecystitis, 3) chronic cholecystitis, 4) acute on chronic cholecystitis, or 5) gangrenous cholecystitis. There was a complete pathologic agreement between grade 5 and a pathologic diagnosis of gangrene, as no other grade of GB had a diagnosis of gangrene. Acute and acute on chronic cholecystitis were found in grades 3-5, while chronic cholecystitis covered the spread with at least one of each grade (Figure 3). The complete demographic and clinical characteristics

ACCEPTED MANUSCRIPT 7

of the 50 patients associated with the 50 respective intraoperative photographs across the 5 image grades are shown in Tables 2-4.

RI PT

DISCUSSION

The need for a surgical scoring system to aid in the prediction of operative difficulty is imperative. Pragmatically, to be able to predict with accuracy and reliability the elevated

SC

difficulty of a LC may assist a surgeon in the decision to convert to an open operation sooner, or call for more experienced hands. Studies have demonstrated that the removal of an acutely

M AN U

inflamed GB can be extremely difficult and should be considered an advanced laparoscopic procedure that may require additional expertise.6 However, the definition and stratification of inflammation remains elusive.

When caring for a patient with acute cholecystitis, there is currently a lack of a uniform

TE D

grading system of disease severity.7 In an effort to standardize disease severity, the American Association for the Surgery of Trauma has developed a grading system that grades the anatomic severity of inflammation in multiple emergency general surgery conditions. This scale ranges

EP

from 1 to 5, each grade representing an escalation from mild to severe, widespread disease.8 While this scale has been validated in certain diseases such as colonic diverticulitis,9 no current

AC C

validation study for acute cholecystitis has been performed. In addition, the impact of each grade of each disease on surgical difficulty and patient outcomes has yet to be verified. A number of predictive factors of LC difficulty have been verified in previous studies.10

Preoperative factors such as male sex, diabetes mellitus, previous surgery, history of cholecystitis, white blood cell count, GB wall thickness, pericholecystic fluid, and an elevated C reactive protein have all been demonstrated to be predictive factors for surgical difficulty and

ACCEPTED MANUSCRIPT 8

open conversion.11-13 Multiple grading scales of LC difficulty have thus been developed in the past based on such risk factors. Most of these scoring systems, however, only utilize preoperative findings, and are derived from single institutions with limited power.14-16

RI PT

A recent risk score was developed by Soltes et al. (2014) to predict the difficulty of LC. Based on history, physical examination, and abdominal ultrasonography measures, this risk score utilizes five levels of difficulty with significant differences in operative time, difficulty, and open

SC

conversion rates.17 Their study, however, only assessed elective cases and did not take into

consideration any intraoperative findings. Another group, Hiroto et al. (2006), proposed using

M AN U

the Tokyo Guidelines to stratify levels of GB inflammation, but this time it was used for acute cholecystitis.18 Again, however, the authors used preoperative findings to grade each class. While the previous scoring methods were based on preoperative data, the Surgical Apgar Score was developed to aid in the prediction of morbidity and mortality after a general or

TE D

vascular surgery utilizing intraoperative data.19 Using the variables of estimated blood loss, lowest mean arterial pressure, and lowest heart rate, this scale has been useful as a postoperative predictor of morbidity20; however, these measures compiled postoperatively are not useful for

EP

intraoperative decision-making. In an effort to predict LC difficulty, Sugure et al. (2015) developed a new grading system based on intraoperative findings. With 10 different grades, this

AC C

scale has been found to be complex, as it includes multiple factors such as patient body mass index, adhesion presence, GB distention, and time to dissect out the critical view.4 In addition to complexity, the scale has not been validated, and the quality of adhesions is not specified. Most grading scales which have been developed are used to predict the risk of conversion

to an open cholecystectomy.4 There is a paucity in the literature of scoring systems to predict other metrics such as hospital length of stay, iatrogenic injury, and total operative time. As acute

ACCEPTED MANUSCRIPT 9

care surgery is currently the most expensive cause of emergency hospitalization in the US, with an annual cost of over $28 billion,21 the avoidance of additional cost is of utmost importance. Additional experience and training in laparoscopy in other surgical operations has demonstrated

RI PT

significant differences in length of stay, intensive care unit admission, and complication rates,22 which can both improve both patient outcomes and lower cost. To be able to better predict the degree of difficulty of an operation, a surgeon may be able to make a more informed decision,

SC

knowing when to convert to an open operation or call for more experienced aid.

We created a grading scale of GB inflammation with an ICC of 0.804, demonstrating

M AN U

strong agreement amongst raters utilizing this scale. While this study was not powered to demonstrate significant differences between grades, trends are appreciated amongst the grades in Tables 3 and 4. Preoperative findings such as a thickened GB wall and pericholecystic fluid are both highest in grade 5 of disease. Operating room time, bile leak rate, length of stay, and

TE D

conversion rate are all highest at grade 5 as well. In addition, Figure 3 shows a trend in pathologic findings between grades. No grade 1 or 2 GBs were found to be acute or acute on chronic cholecystitis. Gangrenous cholecystitis was only found in GBs labeled as grade 5. We

EP

hypothesize that a such a grade-five gallbladder should have a longer operative time, increased operative difficulty, and increased post-operative complication rate compared to lower grades;

AC C

however, this study was not powered to demonstrate such a significant difference. Future work will require additional patients to appropriately power a study that can evaluate potential differences in postoperative complications and outcomes between grades. In addition to the prediction of the difficulty of an LC, our grading system may also help

change the billing practices for surgeons. Current procedural terminology codes are used to communicate uniform information about procedures among different medical providers.23

ACCEPTED MANUSCRIPT 10

Although a “22” modifier can be added to represent increased procedural time or technical difficulty, there is no reliable metric that can corroborate that effort. Other operations, such as a skin grafting, have one code for a certain area covered, and then additional codes can be added as

RI PT

the wound covered increases in size. These extra codes are meant to acknowledge the additional work put forth for the given operation, thus justifying increased compensation. A system with additional codes could also pertain to GBs. A LC is also one of the few laparoscopic cases that is

SC

designated as “basic” for resident case logs that are needed to graduate from residency. However, some of the more difficult, acutely inflamed GBs are often extremely difficult and should be

M AN U

considered an “advanced” case.6 Our grading scale could be utilized to standardize the description of GB inflammation and would help to delineate difficult from straightforward cases. If complexity of case could be documented, then difficult GBs may be logged as “advanced” laparoscopic cases. Another justification for a standardized scale is that operating room time

TE D

utilization is often based on the average time needed to complete the index case. Correlating grades of cholecystitis on outpatient and inpatients in an institution could allow for a more accurate representation of operating room time needs.

EP

A few limitations exist within our study. First, the ICC analysis was based on a retrospective review of still, initial view, intraoperative GB images. A live, intraoperative image

AC C

may demonstrate different qualities and thus could affect the overall grade. Second, this study was derived from a single institution. While 11 different surgeons graded each image, they are all part of the same burn, trauma, and critical care division, which could have affected responses that would have been different at a separate institution. Future work will focus on further reliability assessment and validation of our grading scale and evaluate its relationship with such outcomes as operative time, complication rates, and

ACCEPTED MANUSCRIPT 11

length of stay. We aim to see if differences in such outcomes can be seen between grades or groups of grades with a larger number of patients. In this sense, our five-tiered grading system is our starting point, and we aim to refine this scale if neded over time with additional data. Finally,

RI PT

a multicenter study is required to further evaluate the interrater reliability amongst reviewers from different centers and regions.

SC

CONCLUSION

This study proposes a simple, reliable grading system that characterizes GB complexity based on

M AN U

inflammation and anatomy. The classification system meets the a priori requirements of being simple and easily reproducible among surgeons with the highest level of agreement in those with severe inflammation. This study lays the groundwork to determine if it can be used for risk stratification and to predict patient outcomes, increase resident complex laparoscopic case entry,

AC C

EP

TE D

and improve changes in surgeon reimbursement.

ACCEPTED MANUSCRIPT 12

REFERENCES

6.

7.

8.

9. 10. 11.

12.

13.

14.

15.

16.

RI PT

SC

5.

M AN U

4.

TE D

3.

EP

2.

Everhart JE, Khare M, Hill M, Maurer KR. Prevalence and ethnic differences in gallbladder disease in the United States. Gastroenterology. 1999;117(3):632-639. Csikesz NG, Singla A, Murphy MM, et al. Surgeon volume metrics in laparoscopic cholecystectomy. Digestive diseases and sciences. 2010;55(8):2398-2405. Hussain A. Difficult laparoscopic cholecystectomy: current evidence and strategies of management. Surgical laparoscopy, endoscopy & percutaneous techniques. 2011;21(4):211-217. Sugrue M, Sahebally SM, Ansaloni L, Zielinski MD. Grading operative findings at laparoscopic cholecystectomy- a new scoring system. World journal of emergency surgery : WJES. 2015;10:14. Hallgren KA. Computing Inter-Rater Reliability for Observational Data: An Overview and Tutorial. Tutorials in quantitative methods for psychology. 2012;8(1):23-34. Schwaitzberg SD, Scott DJ, Jones DB, et al. Threefold increased bile duct injury rate is associated with less surgeon experience in an insurance claims database: more rigorous training in biliary surgery may be needed. Surgical endoscopy. 2014;28(11):3068-3073. Crandall ML, Agarwal S, Muskat P, et al. Application of a uniform anatomic grading system to measure disease severity in eight emergency general surgical illnesses. The journal of trauma and acute care surgery. 2014;77(5):705-708. Shafi S, Aboutanos M, Brown CV, et al. Measuring anatomic severity of disease in emergency general surgery. The journal of trauma and acute care surgery. 2014;76(3):884-887. Savage SA, Klekar CS, Priest EL, et al. Validating a new grading scale for emergency general surgery diseases. The Journal of surgical research. 2015;196(2):264-269. Rosen M, Brody F, Ponsky J. Predictive factors for conversion of laparoscopic cholecystectomy. American journal of surgery. 2002;184(3):254-258. Nidoni R, Udachan TV, Sasnur P, et al. Predicting Difficult Laparoscopic Cholecystectomy Based on Clinicoradiological Assessment. Journal of clinical and diagnostic research : JCDR. 2015;9(12):Pc09-12. Kabul Gurbulak E, Gurbulak B, Akgun IE, et al. Prediction of the grade of acute cholecystitis by plasma level of C-reactive protein. Iranian Red Crescent medical journal. 2015;17(4):e28091. Yang TF, Guo L, Wang Q. Evaluation of Preoperative Risk Factor for Converting Laparoscopic to Open Cholecystectomy: A Meta-Analysis. Hepato-gastroenterology. 2014;61(132):958-965. Agrawal N, Singh S, Khichy S. Preoperative Prediction of Difficult Laparoscopic Cholecystectomy: A Scoring Method. Nigerian journal of surgery : official publication of the Nigerian Surgical Research Society. 2015;21(2):130-133. Gupta N, Ranjan G, Arora MP, et al. Validation of a scoring system to predict difficult laparoscopic cholecystectomy. International journal of surgery (London, England). 2013;11(9):1002-1006. Randhawa JS, Pujahari AK. Preoperative prediction of difficult lap chole: a scoring method. The Indian journal of surgery. 2009;71(4):198-201.

AC C

1.

ACCEPTED MANUSCRIPT 13

22.

23.

RI PT

SC

21.

M AN U

20.

TE D

19.

EP

18.

Soltes M, Radonak J. A risk score to predict the difficulty of elective laparoscopic cholecystectomy. Wideochirurgia i inne techniki maloinwazyjne = Videosurgery and other miniinvasive techniques. 2014;9(4):608-612. Hirota M, Takada T, Kawarada Y, et al. Diagnostic criteria and severity assessment of acute cholecystitis: Tokyo Guidelines. Journal of hepato-biliary-pancreatic surgery. 2007;14(1):78-82. Gawande AA, Kwaan MR, Regenbogen SE, et al. An Apgar score for surgery. Journal of the American College of Surgeons. 2007;204(2):201-208. Chandra A, Mangam S, Marzouk D. A review of risk scoring systems utilised in patients undergoing gastrointestinal surgery. Journal of gastrointestinal surgery : official journal of the Society for Surgery of the Alimentary Tract. 2009;13(8):1529-1538. Ogola GO, Gale SC, Haider A, Shafi S. The financial burden of emergency general surgery: National estimates 2010 to 2060. The journal of trauma and acute care surgery. 2015;79(3):444-448. Hsu GP, Morton JM, Jin L, et al. Laparoscopic Roux-en-Y gastric bypass: differences in outcome between attendings and assistants of different training backgrounds. Obesity surgery. 2005;15(8):1104-1110. Leslie-Mazwi TM, Bello JA, Tu R, et al. Current Procedural Terminology: History, Structure, and Relationship to Valuation for the Neuroradiologist. AJNR American journal of neuroradiology. 2016.

AC C

17.

ACCEPTED MANUSCRIPT 14

TABLES

RI PT

Table 1. Parkland Grading Scale for Cholecystitis

Cholecystitis Severity Grade

Description of Severity

Normal appearing gallbladder (“robin’s egg blue”) • No adhesions present • Completely normal gallbladder

2

Minor adhesions at neck, otherwise normal gallbladder • Adhesions restricted to the neck or lower of the gallbladder

3

Presence of ANY of the following: • Hyperemia, pericholecystic fluid, adhesions to the body, distended gallbladder

4

Presence of ANY of the following: • Adhesions obscuring majority of gallbladder • Grade I-III with abnormal liver anatomy, intrahepatic gallbladder, or impacted stone (Mirrizi)

5

Presence of ANY of the following: • Perforation, necrosis, inability to visualize the gallbladder due to adhesions

AC C

EP

TE D

M AN U

SC

1

ACCEPTED MANUSCRIPT 15

Table 2. Demographic Characteristics of the 50 Patients across Grades Grade 1 (n = 8)

Grade 2 (n = 12)

Grade 3 (n = 12)

Grade 4 (n = 9)

Grade 5 (n = 9)

39 (28-43)

33 (29-45)

34 (30-41)

38 (31-43)

48 (41-51)

Female (%)

87.5

100

91.7

66.7

55.6

Hispanic (%)

62.5

83.3

83.3

77.8

77.8

Age*

AC C

EP

TE D

M AN U

SC

*Median (Interquartile range)

RI PT

Demographic Characteristics

ACCEPTED MANUSCRIPT 16

Table 3. Preoperative Clinical Characteristics of the 50 Patients across Grades Grade 2 (n = 12) 0 0 9.4 (8-11) 23.5 (20-36) 24 (17-33) 84.5 (80-120) 0.4 (0.2-0.7)

Grade 3 (n = 12) 16.7 8.33 9.3 (7-11) 49 (23-77) 46 (25-75) 81.5 (101-104) 0.5 (0.2-0.8)

Grade 4 (n = 9) 11.1 0 8.8 (8-11) 37 (26-190) 52 (26-172) 99 (96-122) 0.5 (0.4-1.1)

Grade 5 (n = 9) 44.4 22.2 11.8 (9-14) 23 (20-44) 26 (14-66) 87 (78-94) 0.5 (0.4-1)

RI PT

Preoperative Grade 1 Measures (n = 8) 12.5 % GBW (US) 0 % PCF (US) 8.7 (7-9) WBC* 85 (20-757) AST* 59.5 (20-659) ALT* 114.5 (108-149) ALP* 0.5 (0.3-1.8) TB*

AC C

EP

TE D

M AN U

SC

% GBW, percent with thickened gallbladder wall; US, ultrasound; % PCF, percent with pericholecystic fluid; WBC, white blood cell count; , AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALP, alkaline phosphatase; TB, total bilirubin *Median (Interquartile range)

ACCEPTED MANUSCRIPT 17

Table 4. Perioperative Clinical Characteristics of the 50 Patients across Grades Grade 1 (n = 8) 64.5 (52-82) 0 1.8 (1-2) 0

Grade 2 (n = 12) 62.5 (55-74) 0 1.8 (1-3) 0

Grade 4 Grade 5 (n = 9) (n = 9) 73.0 (72-77) 93.0 (72-100) 0 11.1 2 (2-5) 2.5 (1-3) 0 11.1

AC C

EP

TE D

M AN U

SC

OR, operating room; min, minutes; OC, open conversion, LOS, length of stay *Median (Interquartile range)

Grade 3 (n = 12) 73.5 (52-92) 0 1.5 (1-2) 0

RI PT

Perioperative Measures OR Time (min)* % OC LOS (days)* Bile Leak %

ACCEPTED MANUSCRIPT 18

FIGURE LEGEND

Figure 1: Grade Examples

RI PT

Figure 2: Intraclass Correlation Amongst Raters. Intraoperative images of 50 gallbladders were arranged randomly. Each rater graded each image. The high reliability of the grading

AC C

EP

TE D

M AN U

Figure 3: Pathology Diagnosis Per Grade

SC

system is demonstrated by the near-identical trend for all 11 raters.

AC C

EP

TE

D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE

D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Disclosure Information

AC C

EP

TE D

M AN U

SC

RI PT

The authors have no conflicts of interest to disclose.